Memory and other cognitive functions reside in part in the pattern and strength of synaptic connections between neurons. Understanding the molecular determinants of synaptic strength has been a longstanding goal of neuroscience, and advances in this field stand to influence our understanding of virtually every neurological disorder from Autism to Alzheimer?s disease. Over the past decade biochemical and conventional molecular and genetic approaches have begun to piece together how interactions between neurotransmitter receptors and other synaptic proteins regulate and control synaptic strength and plasticity, but a major limitation is that there is little or no structural information about how proteins are arranged into signaling complexes at the synapse. Many signaling molecules can only interact with immediately adjacent proteins, and this localization may itself be regulated by experience. Understanding how functional signaling complexes are generated and how they in turn regulate synaptic strength thus requires that we probe the spatial arrangements of proteins within the postsynaptic density (PSD). Conventional approaches do not have sufficient resolution to allow the position of synaptic proteins to be mapped within these tiny (< 1 im) synaptic structures. Here I propose to develop tools to map the spatial arrangements of individual synaptic proteins (such as glutamate receptors) within the PSD, and to determine how these spatial arrangements are influenced by synaptic plasticity, using super resolution light microscopy. By mapping the relative positions of many different proteins within the postsynaptic membrane and PSD we will be able to generate a 3 dimensional model of the protein lattices that comprise the postsynaptic side of the synapse. This method has the promise to put a vast array of biochemical and molecular data on protein-protein interactions into a structural context that is essential for its interpretation, and will add a powerful new tool to the analysis of synaptic function.[unreadable] [unreadable]